4D Flow Cardiac MR in Primary Mitral Regurgitation

Abstract

Background: Four-dimensional-flow cardiac MR (4DF-MR) offers advantages in primary mitral regurgitation. The relationship between 4DF-MR-derived mitral regurgitant volume (MR-Rvol) and the post-operative left ventricular (LV) reverse remodeling has not yet been established.

Purpose: To ascertain if the 4DF-MR-derived MR-Rvol correlates with the LV reverse remodeling in primary mitral regurgitation.

Study type: Prospective, single-center, two arm, interventional vs. nonintervention observational study.

Population: Forty-four patients (male N = 30; median age 68 [59-75]) with at least moderate primary mitral regurgitation; either awaiting mitral valve surgery (repair [MVr], replacement [MVR]) or undergoing "watchful waiting" (WW).

Field strength/sequence: 5 T/Balanced steady-state free precession (bSSFP) sequence/Phase contrast imaging/Multishot echo-planar imaging pulse sequence (five shots).

Assessment: Patients underwent transthoracic echocardiography (TTE), phase-contrast MR (PMRI), 4DF-MR and 6-minute walk test (6MWT) at baseline, and a follow-up PMRI and 6MWT at 6 months. MR-Rvol was quantified by PMRI, 4DF-MR, and TTE by one observer. The pre-operative MR-Rvol was correlated with the post-operative decrease in the LV end-diastolic volume index (LVEDVi).

Statistical tests: Included Student t-test/Mann-Whitney test/Fisher's exact test, Bland-Altman plots, linear regression analysis and receiver operating characteristic curves. Statistical significance was defined as P < 0.05.

Results: While Bland-Altman plots demonstrated similar bias between all the modalities, the limits of agreement were narrower between 4DF-MR and PMRI (bias 15; limits of agreement -36 mL to 65 mL), than between 4DF-MR and TTE (bias -8; limits of agreement -106 mL to 90 mL) and PMRI and TTE (bias -23; limits of agreement -105 mL to 59 mL). Linear regression analysis demonstrated a significant association between the MR-Rvol and the post-operative decrease in the LVEDVi, when the MR-Rvol was quantified by PMRI and 4DF-MR, but not by TTE (P = 0.73). 4DF-MR demonstrated the best diagnostic performance for reduction in the post-operative LVEDVi with the largest area under the curve (4DF-MR 0.83; vs. PMRI 0.78; and TTE 0.51; P = 0.89).

Data conclusion: This study demonstrates the potential clinical utility of 4DF-MR in the assessment of primary mitral regurgitation.

Evidence level: 2 TECHNICAL EFFICACY: Stage 5.

Keywords: 4D flow; cardiac magnetic resonance; left ventricular reverse remodeling; mitral valve surgery; primary mitral regurgitation; transthoracic echocardiography.